Title of article
A Highly Parallel and Scalable Motion Estimation Algorithm with GPU for HEVC
Author/Authors
Xue, Yun-gang School of Computer - National University of Defense Technology , China , Su,Hua-you School of Computer - National University of Defense Technology , China , Ren,Ju School of Computer - National University of Defense Technology , China , Wen, Mei School of Computer - National University of Defense Technology , China , Zhang,Chun-yuan School of Computer - National University of Defense Technology , China , Xiao, Li-quan School of Computer - National University of Defense Technology , China
Pages
16
From page
1
To page
16
Abstract
We propose a highly parallel and scalable motion estimation algorithm, named multilevel resolution motion estimation (MLRME for short), by combining the advantages of local full search and downsampling. By subsampling a video frame, a large amount of computation is saved. While using the local full-search method, it can exploit massive parallelism and make full use of the powerful modern many-core accelerators, such as GPU and Intel Xeon Phi. We implanted the proposed MLRME into HM12.0, and the experimental results showed that the encoding quality of the MLRME method is close to that of the fast motion estimation in HEVC, which declines by less than 1.5%. We also implemented the MLRME with CUDA, which obtained 30–60x speed-up compared to the serial algorithm on single CPU. Specifically, the parallel implementation of MLRME on a GTX 460 GPU can meet the real-time coding requirement with about 25 fps for the video format, while, for , the performance is more than 100 fps.
Keywords
Algorithm , Parallel , Scalable Motion , GPU , HEVC
Journal title
Scientific Programming
Serial Year
2017
Full Text URL
Record number
2607835
Link To Document